Continuous Entity Behaviors CEB and Human Intelligence HUMINT in OneSAF

1 / 19
About This Presentation
Title:

Continuous Entity Behaviors CEB and Human Intelligence HUMINT in OneSAF

Description:

that may be used to control the actions of entities (as opposed to units) ... Humans often exhibit instinctive responses to certain kinds of events in group ... –

Number of Views:186
Avg rating:3.0/5.0
Slides: 20
Provided by: david128
Category:

less

Transcript and Presenter's Notes

Title: Continuous Entity Behaviors CEB and Human Intelligence HUMINT in OneSAF


1
Continuous Entity Behaviors (CEB) and Human
Intelligence (HUMINT) in OneSAF
David Nash, Ph.D.
2
Continuous Entity Behaviors
  • The name is descriptive of the capability
  • It refers to a set of OneSAF behaviors
  • that may be used to control the actions of
    entities (as opposed to units)
  • which operate continuously whenever the entity
    is not performing an assigned (directed)
    behavior.
  • Objectives
  • Provide schedulable, randomized behaviors to
    represent the asynchronous activities of
    lifeforms and vehicles.
  • Allow for manual or semi-automatic specification
    of behavior locus.
  • Manual "Use this specific building as your
    workplace."
  • Semi-automatic System identifies an appropriate
    location within a specified bounding area
  • User is able to specify by data the building
    functions and feature types to select for the
    performance of certain kinds of activities (e.g.
    workplace, shopping, or recreation areas)

3
Types of CEB implemented
  • Go to place Move to a particular location or
    area.
  • Optionally, perform "milling" activities at that
    location that are place-specific. For example,
    within a building an entity will periodically
    move to a different location within the room, as
    if pacing, or walking to a file cabinet, etc.
  • Travel roads - vehicles travel random patterns
    within a road network constrained to a
    two-dimensional space. This models the movement
    of delivery trucks, patrolling security vehicles,
    etc.
  • Loiter - Entities move about periodically within
    a designated area. This may be used to model the
    activities of children playing in a park,
    shoppers browsing an outdoor mall, etc.

4
The Scheduled task trigger type
  • CEB are used to perform activities on a periodic
    basis.
  • The Scheduled task trigger type is responsible
    for execution of CEB at the appropriate time.
  • Scheduled trigger inputs
  • Day of week
  • Time of day
  • Duration (possibly infinite)
  • Multiple CEB may be scheduledwithin the same
    phase to representactivities during multiple
    days ofthe week.

5
Future work tools interface
  • A partial implementation of an interface to
    create schedule "classes" has been developed that
    will make it easier to assign CEB to large groups
    of entities
  • "Within this specified area give me 50 day
    workers, 30 farmers, and 20 street people."
  • The schedule classes day worker, farmer, and
    street person have associated sets of scheduled
    behaviors to represent their habitual activities,
    e.g. for day worker
  • Go to work at 0800 30 minutes
  • Got to lunch at 1200 60 minutes, return at 1400
    30 minutes
  • Go home at 1700 30 minutes

6
Idle behaviors
  • Idle behaviors are executed whenever an entity
    is not responding to an event.
  • Noncombatant idle behaviors include
  • Navigate building perimeter
  • Follow road feature
  • Move to and enter building
  • Select and move to random point

7
Reactions to events
  • Humans often exhibit instinctive responses to
    certain kinds of events in group settings, e.g.
    detonations
  • Entities may be composed with an agent that
    listens for particular events of interest.
  • Upon detection of an event, the agent causes a
    reactive behavior to be performed.
  • The mapping of reactive behavior to stimulating
    event is data-driven, and thus configurable prior
    to run time.

8
Elements of a typical urban scenario
NC's pursue various idle cycle behaviors within AO
An event occurs entities within range respond
As reaction durations elapse, affected entities
return to idle behaviors
9
Example scenarios supported by idle / reactive
behaviors
Plaza bomber
Humanitarian aid
Protest
10
Demonstration
11
OneSAF representation of HUMINT capabilities
12
Representing HUMINT
  • Lifeform entities (e.g. individual combatants,
    civilians) may be configured to possess
    information "nuggets" that are either
  • Free-format
  • Descriptive of an activity (task)
  • Such configuration may take place before or
    during scenario execution
  • Both formats include attributes to indicate
    subject, content, time of receipt, source, and
    classification
  • The HUMINT tab of the Intel Battle Command Frame
    provides a graphical interface that exposes
    nugget contents to human role-players / analysts

13
The Intel Battle Command Frame
14
Types of HUMINT-generated information
  • Free-format Text without any particular
    structure. The preceding slide is an example of
    this type.
  • Task oriented Interview subjects that are
    participants in a future orderable behavior
  • If information transfer is partial, it is
    expressed using these increments
  • Free format
  • WHO The interview subject
  • WHAT The title of the nugget
  • WHEN The nugget timestamp
  • DETAILS The content
  • Task-oriented
  • WHO The ordered actor
  • WHAT The title of the behavior (e.g. Construct
    HVIED)
  • WHEN The behavior trigger (at time, on command,
    or completion of previous
  • DETAILS Input parameters peculiar to the ordered
    behavior

15
Transmission of HUMINT
  • Information may be transferred from one lifeform
    to another by using the Conduct Interview
    orderable behavior
  • The ordered actor approaches the interview
    subject, and after an appropriate time delay,
    information flows from the subject to the
    interviewer.
  • The time needed to conduct the interview, as well
    as the subject's inclination to be forthcoming is
    affected by
  • Side relationship between subject/interviewer
    sides
  • Level of cooperation between subject/interviewer
    sides

16
Attenuating the flow of information
  • Side relationship A discrete characterization of
    how entities from each side relate to each other.
    Side relationships need not be reciprocal.
  • Cooperation A sliding-scale characterization of
    entities' willingness to divulge information.

17
Attenuating the flow of information
  • A randomly generated value is compared to
    data-driven thresholds to determine the
    increments of information that are transferred.
  • Side relationship between interview subject and
    interviewer determines the applicable band of
    thresholds the random value determines which
    increments will be conveyed.

18
Influencing cooperation
  • Certain activities change the willingness of
    sides to cooperate
  • Civil-military operations - Humanitarian relief
    operations
  • Psychological operations - Fires and effects
  • Coercion - Security
  • The effects of these activities are represented
    by the orderable behavior Influence Cooperation
  • Ordered actor occupies the area within which
    influence is desired to be imposed after an
    appropriate delay, the effect on cooperation
    between the side performing the activity and the
    sides of entities transiting the area of
    influence is modified
  • The intensity of the effect is
  • Proportional to the number of entities performing
    the action and the number of influenced entities.
  • Inversely proportional to the size of the area of
    influence.

19
Demonstration
Write a Comment
User Comments (0)
About PowerShow.com